Hello, my name is Enri and I served as the solutions architect for Li-Cycle, a company nestled in the heart of Rochester with visions of a greener future. Li-Cycle shreds raw battery packs to produce black mass powder - a raw material conglomerate of precious metals like Nickel, Cobalt, Copper, and Graphite that can be repurposed to remake batteries.
Before we go any further, due to the extremely litigious nature of Li-Cycle and its leadership, I have been advised to clarify that everything stated hereafter must be read with the phrase “allegedly” at the beginning of every sentence. Everything you are about to read is just my opinion. I come to you biased — not against progress, but against deception and environmental negligence. And if you're an investor or supporter of Li-Cycle, know that your perspective is shaped by biases too. However, recognizing our biases doesn't detract from the truth; it sharpens our focus on it.
Join me, not as an engineer or an insider, but as a concerned citizen, as we peel back the layers of what I've come to call the 'fool’s gold' of Li-Cycle. This is more than a story of one company; it's a window into the pervasive issues threatening the very core of New York's manufacturing spirit and beyond.
Background
My intention for this article is to reach the public at large, including those that may not work in the realm of software development technology as I do. Therefore, it is important that we first dive into some fundamental understandings regarding the work I did while at Li-Cycle and similar work I have done in the past.
My background is in industrial automation and I have exclusively worked in the manufacturing sector. During my time in university, I was primarily trained in materials science (which is basically chemistry research) as well as how to install, validate, and modify various physical machinery used in the manufacturing of semiconductor products (electronic chips).
While in university, I quickly learned that the future of manufacturing resided primarily in learning to build software that controlled the machinery and robotics rather than building the physical machines/robots. Unfortunately, few jobs, if any, exist in New York state where you get paid a sustainable living wage to build, operate and maintain industrial machinery.
In total, combining my university and full-time industry experience, I have had the privilege of working just about every job imaginable in manufacturing. I have worked in the classic white lab coat (also called bunny suits) doing chemistry research, I have worked to publish peer-reviewed journals, I have installed/operated/maintained machines small and large, I have worked in the environmental health & safety division, I have worked in procurement & finance overseeing acquisition of raw materials and machinery capital, I have developed various software applications that collect/store/visualize/analyze data generating by machinery during production, and I have installed/managed traditional IT equipment like servers.
Suffice to say that the only thing left on my bucket list to do as a job within manufacturing is become forklift certified. Maybe one day, a fella can hope.
It used to be the case that the typical career growth trajectory for manufacturing professionals would be to start working on the shop floor as an operator/technician/mechanic, gain seniority to become a shift supervisor, attend trade school or engineering school to perform more advanced functions like designing & research, prove yourself through a series of projects to try to perfect your craft, and gain enough trust to eventually lead/manage a team of your own. This upward mobility was plentiful in the 1970s to early 2000s as the number of jobs were directly tied to the need for manual operation of the machines being installed.
Software automation quickly dissolved the need to be on the factory floor to get exposure to manufacturing operations thanks to the use of virtual factory simulation, in addition to exponentially reducing the number of staff it takes to operate/maintain machinery.
This fundamentally shocked the industry, leaving it with no time to digest the tidal wave of change coming about, unfortunately resulting in unprecedented job loss due to automation coupled with unresponsive social systems to adapt to the change. To clarify the position I held - a solutions architect is the epitome of the software development profession. It is akin to a civil engineer who earns their state P.E. license granting them legal authority to sign off on building construction blueprints.
Environmental, Social, and Governance (ESG) Standards
Software automation allowed society to consume at exponentially faster rates than it ever was before. The paradox of efficiency kicked in immediately, meaning that humans are consuming raw materials faster than the raw materials can regenerate.
This naturally led to questions arising about how to make the materials we consume renewable so that they may be quickly reharvested again for consumption.
Thus ushering in the era of ESG - Environmental, Social, and Governance criteria used to scrutinize investments and the purposes they served. At face value, ESG seems innocent enough. Its core ethos advocates for “putting your money where your mouth is” by choosing to only fund companies that can demonstrate they have a low carbon footprint, have progressive labor practices, do not engage in business with nations that commit human rights violations, have diverse board composition, transparent executive compensation commensurate with company & market performance, and give back to the communities they serve.
Here’s the catch - it is extremely difficult, if not impossible to quantify the “Social” and “Governance” parts of ESG and correlate those values to make claims as to whether such organizations actually are in fact making positive changes to society. The “Environmental” criteria from ESG is something engineers like myself can easily quantify and validate the quality of the source data but the correlations we can make from it may not always be so straightforward.
Here’s a specific example: we can easily measure the amount of pollution generated during manufacturing processes by logging data on how much CO2 was generated, how much volume of water was consumed, the watts of electrical power used, the volume/weight of gasoline or diesel fuel consumed, and the like.
Furthermore, many of the ESG standards used to determine which organizations receive funds were fundamentally flawed in that the bodies mandating the criteria were not democratically elected and were far removed from actually implementing the standards. This always leads to a disconnect from reality between what theoretically sounds good on paper versus how it actually works in the real world.
Fundamentally, one of the many reasons ESG standards as an investment criteria failed is because they said nothing about open sourcing the projects & work companies create. This means that ESG says nothing about things that actually matter, such as mandating companies to publicly release the technology they develop for others to freely use or mandating companies to assume fiduciary responsibility for the wellbeing of the communities they operate in.
Companies of all sizes that received ESG investments have repeatedly shown themselves to do worse in all 3 of the ESG criteria than companies that do not abide by ESG standards. This means that if you take 2 companies that operate in the same exact sector and offer similar products/services, the company without ESG criteria will have a lower carbon footprint than those that brand themselves as abiding by ESG criteria. The stranger paradox is that companies who proclaim to abide by ESG criteria also show higher levels of corruption and are generally less efficient at operating than their non-ESG counterparts.
The takeaway here is that reducing environmental pollution, preventing corruption, and mutually sharing in the bounty reaped from collaboration are noble causes. These causes are absolutely worth fighting for and do need to be fiercely fought for. However, the current ESG standards criteria are absolutely setting everyone up for failure. ESG has primarily been used as a predatory tactic to funnel investor money into organizations that claim to be advocates for prosocial change, when in reality they are wolves hiding in sheep’s clothing.
Enter Li-Cycle.
The Li-Cycle Equation
Now that you have the whole picture of where manufacturing was, where it is today, where it’s going, and the need for renewable materials, how does Li-Cycle fit into this whole charade?
From the get-go, Li-Cycle quickly took on the public mantle of branding itself as an ESG compliant organization. At the time of Li-Cycle’s inception, President Biden assumed office, initiating a series of legislation and discussion around climate change, namely the reduction of greenhouse gas emissions. This initiative, coupled with other politicians advocating for the “Green New Deal” provided significant funds for various renewable energy sector companies - namely photovoltaics (solar panels) and battery recyclers.
The problem is that this legislation assumes that we can already easily track all inputs and outputs using technology that frankly is not there yet for the reasons I have described so far. A true circular economy necessitates that you can monitor inputs and outputs at every point in the system and that each system can easily & freely share data between each other.
You may have also heard of this free exchange of data between systems and individuals as “The Internet of Things (IoT)”. In industrial automation, the buzzword used to describe the free exchange of data amongst machines, from machines to humans, and humans to humans, is called “The Industrial Internet of Things (IIoT)”. Unfortunately, manufacturing is plagued by proprietary technology intentionally designed to NOT be interoperable with other vendors - both at the physical machine level and at the software level.
There is virtually no diversity in products and vendors within manufacturing. A small handful of special interest groups have a chokehold on manufacturing, leading to manufacturing colloquially being known as “the good old boys club”. To be a vendor that provides software and/or hardware services to manufacturers, you must agree to be bullied and harassed by what I call the manufacturing mafia, namely equipment vendors like Emerson, Siemens, Rockwell, Schneider Electric, and their respective affiliates.
Nepotism and Cronyism
At this point, you might be tempted to feel pity and provide grace for a newcomer like Li-Cycle who has to battle against this tide. To some degree, you would be justified in feeling sympathy for them. However, for reasons you will soon see, this sympathy will naturally be short lived.
One of the biggest mistakes manufacturers make when building out a factory is choosing 1 vendor to handle all things related to installing the necessary machinery and validating the quality of this work. Depending upon the specific case, using 1 vendor to conduct the work may be a good idea. However, when it comes to installing equipment on a bare naked factory, relying on just 1 vendor is a terrible idea. Where it gets even worse for Li-Cycle is their tendency to employ family members / relatives to handle the work.
This pattern of cronyism is nothing new to Li-Cycle - as clearly detailed in an incredibly accurate report done by Blue Orca Capital (to which I have no affiliation with). For all of their “spoke” factories, which were intended to provide feed stock for the “hub” factory, Li-Cycle decided not to hold any sort of public bidding/RFP to execute the various functions required for installing machinery for their spoke factories.
It is important to note that there are several job functions required to successfully install machinery - electricians to run telecomm & power utilities, HVAC & plumbers to run air & water utilities, software developers to build various applications necessary to monitor the status of the factory operations, network engineers to securely connect & configure the machine controllers to a distributed network to allow for data collection, data engineers to build robust data pipelines to stream data from the machines into various repositories, cybersecurity engineers to secure these data repositories, IT engineers to install servers and implement monitoring tools to oversee the server health, site reliability engineers to install and monitor redundant data centers during the inevitable occurrence of failures, safety engineers to monitor chemical hazards during and after machine installation/operation, mechanical engineers to verify the structural integrity of the machinery and building holding the machinery, and electrical engineers to validate that the electronics operating said machinery have several redundant safety measures as well as operate to specification.
Li-Cycle actually had a team of engineers at its inception who warned the company that it would be impossible to get all of these needs met with just 1 vendor. But sure enough, Li-Cycle ignored their heed and decided to hire Co-Founder Tim Johnston’s personal friend, Chris Bisson of Bisson Innovations, to carry out all of this work. With no one available internally to validate the work that Bisson Innovations did, Li-Cycle would quickly learn how badly it messed up.
Rookie Mistakes - Shoddy Process Control Systems
To say that Bisson Innovations did shoddy work would be an understatement. Bisson Innovations markets themselves as a “full service shop”, claiming that it can execute all the functions I mentioned earlier - a feat that even giants like Siemens and Emerson hesitate to make. Large OEMs like Siemens understand the complexity involved, which is why they typically contract this kind of work to specialty shops. Two-ish years later and many tens of millions of dollars down the drain, Li-Cycle was stuck with what I can only describe as a college freshman’s first attempt at making a process control system.
Each spoke factory is running off of a singular Dell desktop computer, which hosts all the SCADA software running the plant as well as stores all process control data locally as flat files on generic hard drives meant to be used for consumer use.
One would hope that these pseudo-SCADA desktops would be connected to some sort of uninterruptible power supply (UPS). Nope. In fact, the Rochester spoke was the very first spoke factory to officially have a UPS and a spare hard drive after I visited the local Best Buy. The funniest part about this Best Buy trip was I noticed the new genius IQ levels of Bisson Innovations.
This systems integrator tightly coupled the HMI screens (screen dependency) whereby moving one screen immediately bricks all other screens. Furthermore, although just 1 HMI screen would have issues, it would automatically brick the applications running on all the other HMI, leading me to believe that these HMI were storing app data locally. It became very clear to me that modular design & following best practices were out the window.
But wait, it gets worse! I came to find out that these Dell desktops had failed multiple times since the time they had been installed - as one should expect when you try running your entire factory from a singular computer meant for basic web browsing and emails. When I tried installing an endpoint backup utility on the workstation, it turns out that no one at Li-Cycle had any of the administrative credentials required to do things like add/delete/modify programs. My dream had come true! I was literally witnessing tyranny (or as some of you may called it “vendor lock-in”) by a systems integrator vendor!
And guess which OEM vendor Bisson Innovations is affiliated with? That’s right! It’s Rockwell! Because of course it is.
Of all the OEMs that are likely to rob a customer blind, it is well known that Rockwell has over 9000% chance to be the first ones to do so. But it doesn’t stop there. It’s one thing to NOT have admin credentials to any of your “pseudo-SCADA” desktop. This problem was literally occurring at every spoke factory. EVEN WORSE - the systems integrator refused to hand off the usual deliverables when a project is turned over to the customer. Control narratives, data architecture diagrams, software bill of materials, network topology diagram. Nothing.
You would think with all of this lackluster experience from the integrator, Li-Cycle would not do business with them anymore. Unfortunately, yet again, you would be wrong. In classic Li-Cycle double-down fashion, Li-Cycle chose to continue giving Bisson Innovations no contest contracts to do all aspects of systems integration for its spoke factories in Europe.
Theoretically, a manufacturer like Li-Cycle could be run using the same cookie-cutter SCADA and HMI configuration since there’s only 1 product, black mass. However, not all shredders are made equally. There is natural variation due to the uniformity of the shredding and drying machines as well as the physical layout of each factory. Despite this reality, it definitely did not stop the integrator from trying to do just this. See, Li-Cycle does not even have a real SCADA system - what they have is a basic P&ID visual with no sort of alarming/notification/remote execution but the P&ID does not necessarily align with the individual spoke factory.
It did not take me long to find all of this out and begin the process of escalating it to the necessary leaders. And here’s where we get to some of my favorite parts in this entire convoluted story.
Li-Cycle Leadership
Although Li-Cycle did technically have job titles with leadership implied in the name, leadership traits were not present in any of them. One of my fondest memories is seeing the VP of IT having to obtain formal written approval to authorize a purchase of less than $5 to install some remote endpoint access tool.
Remember folks, a corporate job title does NOT automatically make you a leader. Micromanagement and technical incompetence ran amok. Prior to the current VP of IT I worked with, Li-Cycle actually had another VP of IT. It seems that Li-Cycle churns & burns through high-ranking staff for some reason, kinda like there’s something severely wrong going on internally and people leaving because they don’t want their names attached to it.
For those that don’t know what a VP of IT does, here’s an overview:
Work alongside the CTO (which Li-Cycle did have) to procure various technology services to meet business needs.
Expected to be a subject matter expert in all matters surrounding information technology, including but not limited to network engineering, data backend engineering, infrastructure equipment, site reliability, application development, and cybersecurity engineering.
If the knowledgebase to execute on procuring these services does not exist internally, the VP of IT must hire internal full-time staff as well as procure external contracted consultants.
Li-Cycle’s first VP of IT made it very clear they were completely out of place by spending tens of millions of dollars purchasing the SageMaker machine learning platform from AWS to perform complex modeling & analytics using their process control data. There’s just 1 problem - Li-Cycle has zero data architecture and what little data pipelines they had at their Arizona factory was poorly built using misconfigured OPC infrastructure.
This is what we call “garbage in, garbage out”. AWS SageMaker (any ML framework for that matter) necessitates that all your data pipelines are feeding cleanly formatted, validated, accurate, and contextualized data in order to execute its machine learning functionality. Building the data pipelines and ensuring the 3 other facets I mentioned are the hardest parts of working with data. This original VP of IT knew this, yet decided to go down this path anyway.
Even worse is that when I arrived and confronted the current VP of IT, he too did nothing to stop this senseless hemorrhaging of money. It was only after a few months following the construction shutdown that Li-Cycle ceased its services with AWS because they literally ran out of funds to keep it going & not because they listened to their subject matter expert.
I spent countless hours with the current VP of IT through 1:1 meetings, writing educational material, and explaining technical concepts only to be greeted with complete indifference. It became evident to me that I was dealing with a VP of IT who did not understand basic computer science fundamentals like building an application and the multitude of factors to consider to persist its data.
Failure to stop hemorrhaging money on useless initiatives did not stop there. Prior to my arrival, Li-Cycle had decided to single source contract IBM Maximo to meet their CMMS needs to perform tasks like plant maintenance, physical asset serialization, and spare parts management without ever doing any sort of discovery internally prior to procuring the Maximo licenses and systems integration work.
To make matters even worse, Li-Cycle decided to purchase Maximo as a SaaS solution. For those familiar with Maximo, they understand that it is complete overkill for a business like Li-Cycle. For starters, Maximo is a completely closed source solution that makes it impossible to export data out of the Maximo environment to use with other systems that might be able to benefit from data regarding maintenance and spare parts inventory. Furthermore, Maximo is best suited for situations where you have very clear & specific use cases, something Li-Cycle did not as it had not even conducted discovery yet coupled with the fact that Maximo requires significant development time to customize the platform to suit said use cases.
Systems like Maximo are also relatively lightweight to host, it makes zero sense for a company like Li-Cycle who intended to achieve a very large scale to utilize a SaaS-exclusive platform. The infrastructure costs alone would be easily mitigated by deploying AT LEAST a hybrid on-prem/SaaS solution, which Maximo can accommodate.
The most bizarre takeaway from this Maximo situation is that Li-Cycle did not have an actual need to use such a complex solution to meet its maintenance tracking and CMMS needs. Yet again, I explained to the VP of IT and to the IT Director that there are much cheaper alternatives that Li-Cycle can procure and/or build, self-host, save millions, all the while being able to readily extend this cheaper option into another robust environment - commercially procured or built.
The Maximo equation does not end here. Li-Cycle’s obsession with contract-consultants and single source contracts extend into its Rochester Hub factory business decisions, which at this time is still no longer under construction. The choice of reviewing the necessary IT services and equipment to support all of the OT needs was never considered despite already pulling the trigger and buying various machinery and SCADA software from big-name OEM Emerson.
Timer servers, network topology architecture, data centers, redundant data storage, access points, telecommunications utilities, and uninterruptible power supply units had never been discussed prior to buying all the process controls and instrumentation for the Rochester Hub factory. Systems integration had never been discussed and it was unclear to me whether Li-Cycle had even done factory acceptance testing of all the machinery it had already procured. In all my years of industrial automation work, I had never seen such gross negligence. It was clear that Li-Cycle never believed it would ever run out of money.
Their accounting and procurement practices left much to be desired. Anyone could pretty much cut a purchase order & paycheck and no questions were ever asked. I tested this theory by submitting a PO for a PLC training station that Li-Cycle frankly did not need primarily because Li-Cycle already had many of the controllers & field devices connected to unused machinery that could be easily repurposed for training. It was quickly approved and the only reason I did not get the training rack is because I was onboarded just before the construction halt occurred, meaning money literally ran out.
Instead of hiring full-time staff internally to carry out traditional IT support services like imaging workstations and installing applications, Li-Cycle’s addiction to contract consultants resulted in spending hundreds of thousands every month to procure such services from 2 separate managed IT services providers. These 2 providers were frequently colliding with each other, as their scopes of work frequently were indistinguishable.
In the short time I worked at Li-Cycle, they could have easily hired 6-12 full-time internal staff who could complete work in a timely manner instead of bickering over which managed IT services provider has jurisdiction over a ticket. The beautiful thing is that these 6-12 internal full-time staff members would have cost Li-Cycle much less and delivered higher quality work than the monthly costs incurred with 2 separate managed IT services vendors who have to split their time between other customers in their books in addition to Li-Cycle.
What really frustrated me the most is that Li-Cycle leadership absolutely felt entitled to receive all communication exclusively in PowerPoint slide decks. Talk about being extremely inefficient. It is reminiscent of the early cult-like days of Amazon and I honestly cannot take anyone seriously who expects all communication to be done exclusively through powerpoint. Some things can be better portrayed through slide decks but many others cannot. I honestly think that Li-Cycle leadership thinks of themselves as some new-age ESG Steve Jobs / Jeff Bezos. Hilarious. Anyone that cannot comprehend material unless it’s presented in a slide deck should have their competency seriously questioned.
More Strange Behavior From Li-Cycle
Sticking to the trend of spending money on large systems and never using them - allow me to introduce you now to eLogger at Li-Cycle. The eLogger product is essentially a glorified Microsoft Forms application. Yet again, Li-Cycle did not need to spend hundreds of thousands of dollars to implement eLogger. The funniest part was that nobody used eLogger and resorted to using Microsoft Forms instead.
The decision to procure all these various software systems and IT infrastructure were clearly being made by people who had no understanding of what the business actually needed. The most egregious offense though was the millions Li-Cycle was spending on its ERP software from Oracle every year.
Yet again, I approached Li-Cycle leadership explaining that their needs are simple and much cheaper & scalable alternatives exist - namely Odoo. Li-Cycle only has 1 product it sells and that product is black mass. This greatly simplifies the need for customized ERP and CRM functionality, all of which are easily accommodated by open source tools like Odoo.
Yet again, Li-Cycle leadership greeted me with indifference.
The clown show does not stop here. Just before I was let go as part of the massive reorganization, Rochester Police showed up to the hub factory confronting Li-Cycle about unauthorized installation & misconfiguration of an LTE fire alarm system installed in the warehouse building at their hub factory facility. Implementing an LTE-based industrial fire alarm system is a normal part of building a factory.
The abnormal part was that Li-Cycle did not follow through the normal procedure for implementing such a system. There’s a whole permit process and network configuration that must take place to ensure that your LTE system does not interfere with other people’s service, that you are paying dues for occupying those airwaves, and that they are not being used for nefarious purposes.
Even I, the architect who is supposed to be included in all the correspondence relating to hardware and software systems pertaining to manufacturing operations at every level, did not know about this. You can imagine what it’s like to see an entire army of cops at your place of work.
Stranger still was the moment I was laid off. I opted to return my company issued laptop & badge card via mail and requested that my possessions be returned via mail as well. Thankfully, before I left, I added in my PO box as my shipping address in addition to my home address. Normally, the company would ship your possessions to you to the mailing address listed on the human capital management (HCM) database. Not at Li-Cycle. The HR department insisted that they show up to my home to return my possessions.
I was vehemently against this and told them to simply return my possessions to the mailing address I provided. I wonder what would have happened if I never made this change. Can you imagine seeing an HR rep showing up unannounced and unwelcomed to your front door, especially after you were just laid off? I have never seen such bizarre behavior from any other company in my entire life.
Similarities to State Initiatives Purporting To Make Manufacturing Great Again
By now, the key takeaways I would expect people to have are:
Want to avoid accountability and prey on people’s philanthropic desire to make the world a better place by investing in recycling initiatives? Go public with a SPAC and brand yourself as an ESG company.
Most people with a Director, VP and C-level job title have no technical competence, are unfit to lead, and are actively causing the ruin of manufacturers worldwide.
This level of systemic incompetence and negligence is by design. Li-Cycle is simply one duck quacking along this path but they are not unique. As public hype continues to spread around ESG being “the next big thing”, we will absolutely see more corrupt organizations grifting around the topic of climate change, a topic that I hold dear & take very seriously.
To make matters worse, manufacturing industry associations like IEEE, ISA, CESMII, and others do absolutely nothing to call out such predatory behaviors both from manufacturers and by OEMs/systems integrators. One would hope that industry associations would serve as watchdogs advocating for open source technology to educate the public and protect against such predatory behavior. How many other companies are out there who are much smaller than Li-Cycle but still need industrial automation services, only to fall prey to the same manipulative tactics from OEMs/integrators/associations selling them overpriced services they do not need?
This has to stop. Manufacturing is being slowly and surely destroyed from the inside-out by the same groups that we expected to protect & serve.
Then we have state sponsored programs like NYSTAR and the Upstate Revitalization Initiative, which have miserably failed to deliver on their promises of creating manufacturing jobs with living wages & technological innovation accessible by all with no barriers to entry.
Performance and Accounting Methods
One of the most surprisingly accurate documents I came across during my time with Li-Cycle was the report written by Blue Orca Capital about their speculation on Li-Cycle’s performance.
It was more than just its accuracy that surprised me, it was that I actually found Ashlin marketing material in one of my backup offices at the Rochester spoke factory. Seemed to me like someone forgot to take away the Ashlin connection after purchasing all the various random swag bag packages stored in the backoffice grey metal cabinet next to the standing desk.
The second thing I found most surprising from the report by Blue Orca Capital was how accurate their financial performance analysis of Li-Cycle was. One of the landmark tools I led the build of was a visual dashboard showing the black mass produced & raw material shredded for every spoke factory that Li-Cycle operated. This data was coming from a controlled inventory-adjusted data backend source using a database connection to their ERP inventory tables.
It was clear to me that Li-Cycle had not been producing as black mass as they had been announcing for quite some time, even long before the whole construction shutdown occurred. The moment this project went live, countless people within the company reached out to get their hands on it. I’m not sure if it’s still active or not today but I do know one thing for sure - that entire application is sitting on a service account on a production server, so any changes (especially deletions) will most assuredly come up in a financial audit report.
I’m sure investors would love to see this “Spoke performance report” visual.
Black Mass Quality
The real evil of the Enron accounting practices used by Li-Cycle are best seen with their product quality. What exactly does product quality mean for a battery recycler? It’s actually a lot easier than you may think. There are 3 main factors when evaluating the quality of the black mass collected after shredding the original batteries:
- The concentration & potency of precious metals (Nickel, Copper, Cobalt) present in the collected black mass.
- The ability to remove moisture from the collected black mass and keep it away as the black mass sits as inventory in the warehouse.
- The ability to completely separate scrap material from the black mass so that the powder can undergo catalysis.
Li-Cycle was misrepresenting their total black mass collected after shredding the spent batteries by conflating the black mass collected value to the total raw material shredded in addition to showing projected future revenue. Overall, the ratio of raw material shredded to black mass collected was incredibly small and warrants an entire conversation around the effectiveness of lithium battery recycling. Is it even true that the process of battery recycling results in less pollution than simply mining new ones? The answer to this question was not so clear to me after following the data.
Li-Cycle had no product quality management. Li-Cycle began conducting quality checks in the last few months. Li-Cycle would conduct a manual inspection on a limited basis by having a skeleton crew of operators scan a handful of black mass already bagged for storage to test for the presence of precious metals. The problem is that this “quality inspection testing” did nothing to test for their potency and Li-Cycle did not take any readings of moisture levels in the black mass.
One of the projects I was working on was to store the test data from these scanners onto a robust & controlled data repository. Once a test reading was completed by an operator using a scanner, it was supposed to be offloaded into a controlled data repository to strictly monitor access to this data. After repeatedly trying to work with the 2 managed IT services vendors Li-Cycle procured, I had run into a brick wall. I was requesting hardened computer workstations be deployed at the site where the operator docks the scanner to allow for offloading of the test data.
Despite imaging and domain joining the workstations, both of the managed IT services vendors neglected to install endpoint management tools on these machines and refused to cooperate in my attempts at creating a database on a hardened server to store the test results. To this day, I can guarantee that the quality engineer is storing test data onto their personal Microsoft Office365 account via their OneDrive repository. After repeated attempts at escalating this issue to Li-Cycle leadership, I was met with blatant indifference and negligence.
The fun does not stop there. See, good quality management is done by performing measurements while the product is being made as well as after the product is made. As I mentioned earlier, Bisson Innovations - the systems integrator responsible for setting up the entire data architecture to contextualize all manufacturing process data - had not done anything to create a data architecture.
This means that no one at Li-Cycle had the ability to actively monitor any readings to tell whether the batteries they were shredding & attempting to catalyze were any good. At best, Li-Cycle was winging their entire operation collecting black mass, hoping that the end buyer did not realize the junk quality before money exchanged hands. It will be interesting to see the studies years from now on defective batteries and to trace their origins to black mass originating from Li-Cycle.
Oddly enough, it seems that Li-Cycle was at least thinking about quality to some extent. They had procured a data historian solution by OSI PI in a futile attempt to do some level of data analysis on their manufacturing process. Again, keep in mind the “garbage in, garbage out” phenomenon discussed earlier.
Li-Cycle has no data architecture, meaning no one (except the systems integrator Bisson Innovations) can make sense of the process data because Bisson Innovations had full control over the naming conventions & hierarchical organization of the data tags. I checked to see if there was some naming convention and pattern to how data from the edge was organized and honestly, it looked like Greek to me.
The odd thing was that the PI system was only being used at 1 of their spoke factories. Come to find out that yet again, there was no deliberation when procuring an entire data historian solution. Li-Cycle single sourced the procurement of OSI PI data historian to a local Rochester-based channel provider to the OSIsoft company, the creators of the OSI PI data historian product. Even stranger was how much Li-Cycle had overspent on license costs and implementation of the OSI PI system.
They purchased about 50,000 tags from PI and spent close to $750k. Bizarrely, the Li-Cycle VP of IT had decided to host the OSI PI data historian solution entirely on the cloud, which was adding significant costs to this already absurd number. These expenses were all unnecessary. The overall throughput of battery shredding is, by nature, VERY VERY slow and for good reason. Assurances must be in place for the battery to be properly grounded and discharged, coupled with the long time it takes to perform a first-pass drying attempt for the recovered black mass powder.
Given the low throughput and low volume of data being generated by the manufacturing process, Li-Cycle would have been better served to building inhouse applications to read the edge data from flat files stored on redundant Network Attached Storage (NAS) racks. I explained just as much to Li-Cycle leadership, including the VP of manufacturing operations, only to be once again met with indifference and inaction.
Oddly enough, even after acquiring the recent $75 million investment, it seems that Li-Cycle could not be bothered to spend a one-time fee of $10k on a NAS rack that could also serve countless other purposes. Instead, they chose to continue hemorrhaging money with their cloud hosted OSI PI implementation that nobody was even using anyway because less than a handful of people even knew how to use the OSI PI data historian product.
Honestly, I don’t blame them because even I struggle using it due to its horrific design. It’s why I try to completely circumvent it by using what little API that OSIsoft has publicly available on programmatically querying data from the backend using open protocols like MQTT. So it seems that once again, Li-Cycle has achieved unnecessary waste nirvana - they have managed to spend about $1 million on a product that nobody used and that nobody stringently audited the procurement of. Just fabulous. Just keep spending money guys, the money printer goes brrrr. That DoE loan will be here soon - said literally every Li-Cycle staffer & member of leadership ever.
Clown show. Shame on the channel provider for selling yet another customer products and services you knew they did not need. I guess ethics and due diligence go out the window when money is pouring in.
Conclusion
So how do we move forward from here? In my opinion, the only way to win this war against corruption and negligence within industrial automation is to support open source projects - both existing ones and the creation of new ones.
The chokehold that proprietary closed source solutions have had on manufacturing have crippled its ability to innovate and serve as the glue that holds families together. Manufacturing jobs used to be the lifeblood of society, providing middle class income, security, long term upward mobility and the ability to build generational wealth. Private ownership of everything, including technological discoveries, only serves to benefit special interest groups at everyone else’s expense.
We must make open source a top priority at every level and discuss the nuance associated with it. There are many practical steps to support the open source movement:
Write your local government representatives (senator, congressman, mayor) and ask them for their stance on the right to repair and support open source projects as an investment vehicle to create jobs. You can check out my YouTube channel where I show you exactly how to do so.
DM me to learn how you can assist me in making a watchdog platform to call out bad actors in the industrial automation space. One person can truly make a difference.
Stay tuned to content I make to learn more about open source and the intimate ways it’s tied to consumer rights, household electronic devices, and industrial automation. Education is important.
I challenge other solutions architects to come forward with your stories. I hear from you every day facing the exact same problems in the organizations you work for. I challenge systems integrators to stop this vicious cycle of shilling proprietary closed source products/services that are clearly NOT meeting customer needs, despite decades of this same old strategy.
The future is open source and that future is bright.
Thank you all for your time and consideration. You could have been doing anything else in the world right now but instead you chose to review my work & for that I am forever grateful, even if we may disagree. Stay tuned and I’ll see you in the next one.
Follow me
› Linktree: https://linktr.ee/opensourceadvocate
› LinkedIn: https://www.linkedin.com/in/enrimarini
› Substack: https://enrimarini.substack.com/
› Twitter: https://twitter.com/@RealEnriMarini
› Medium: https://medium.com/@TheEthicalEngineer
› YouTube: https://www.youtube.com/@EthicsAndEngineering
› DEV Community: https://dev.to/@opensourceadvocate
› TikTok:https://www.tiktok.com/@opensourceadvocate
—
DISCLAIMER: I am not sponsored or influenced in any way, shape, or form by the companies and products mentioned. This is my own original content, with image credits given as appropriate and necessary.
Top comments (0)